ICCV - arXivDaily 期刊&会议

2508.21060 2025-08-29 cs.CV

Multi-View 3D Point Tracking

Frano Rajič, Haofei Xu, Marko Mihajlovic, Siyuan Li, Irem Demir, Emircan Gündoğdu, Lei Ke, Sergey Prokudin, Marc Pollefeys, Siyu Tang

机构 * ETH Zürich（苏黎世联邦理工学院）； Carnegie Mellon University（卡内基梅隆大学）； Balgrist University Hospital（巴尔格里斯大学医院）； Microsoft（微软公司）

Comments ICCV 2025, Oral. Project page: https://ethz-vlg.github.io/mvtracker

2508.21032 2025-08-29 cs.CV

Reusing Computation in Text-to-Image Diffusion for Efficient Generation of Image Sets

Dale Decatur, Thibault Groueix, Wang Yifan, Rana Hanocka, Vladimir Kim, Matheus Gadelha

机构 * University of Chicago（芝加哥大学）； Adobe Research（Adobe研究）

Comments ICCV 2025. Project page: https://ddecatur.github.io/hierarchical-diffusion/

2508.20626 2025-08-29 cs.CV cs.AI

ArtFace: Towards Historical Portrait Face Identification via Model Adaptation

Francois Poh, Anjith George, Sébastien Marcel

机构 * Idiap Research Institute（IDiap研究 institute）

Comments 4 pages, 3 figures. ArtMetrics @ ICCV 2025 (non-archival). Paper page at https://www.idiap.ch/paper/artface/

2508.20505 2025-08-29 cs.CV

Describe, Don't Dictate: Semantic Image Editing with Natural Language Intent

En Ci, Shanyan Guan, Yanhao Ge, Yilin Zhang, Wei Li, Zhenyu Zhang, Jian Yang, Ying Tai

机构 * Nanjing University, China（南京大学）； vivo, China（vivo）

Comments Accepted by ICCV 2025

2508.20488 2025-08-29 cs.CV

Adaptive Dual Uncertainty Optimization: Boosting Monocular 3D Object Detection under Test-Time Shifts

Zixuan Hu, Dongxiao Li, Xinzhu Ma, Shixiang Tang, Xiaotong Li, Wenhan Yang, Ling-Yu Duan

机构 * School of Computer Science, Peking University（北京大学计算机学院）； Peng Cheng Laboratory（鹏城实验室）； The Chinese University of Hong Kong（香港中文大学）

Comments Accepted by ICCV 2025 (Highlight)

2508.20381 2025-08-29 cs.CV

More Reliable Pseudo-labels, Better Performance: A Generalized Approach to Single Positive Multi-label Learning

Luong Tran, Thieu Vo, Anh Nguyen, Sang Dinh, Van Nguyen

机构 * FPT Software AI Center（FPT软件人工智能中心）； National University of Singapore（新加坡国立大学）； University of Liverpool（利物浦大学）； Hanoi University of Science and Technology（河内科学技术大学）

Comments ICCV 2025

2508.20265 2025-08-29 cs.CV cs.LG

Plug-in Feedback Self-adaptive Attention in CLIP for Training-free Open-Vocabulary Segmentation

Zhixiang Chi, Yanan Wu, Li Gu, Huan Liu, Ziqiang Wang, Yang Zhang, Yang Wang, Konstantinos N. Plataniotis

机构 * University of Toronto（多伦多大学）； China Agricultural University（中国农业大学）； Concordia University（Concordia大学）； McMaster University（麦马斯特大学）； Beijing Jiaotong University（北京交通大学）

Comments ICCV 2025, code:https://github.com/chi-chi-zx/FSA

2503.22677 2025-08-29 cs.CV cs.AI cs.LG

DSO: Aligning 3D Generators with Simulation Feedback for Physical Soundness

Ruining Li, Chuanxia Zheng, Christian Rupprecht, Andrea Vedaldi

机构 * Visual Geometry Group, University of Oxford（牛津大学视觉几何组）

Comments Accepted at ICCV 2025 (Highlight). Project page: https://ruiningli.com/dso

2411.00626 2025-08-29 cs.CV

ZIM: Zero-Shot Image Matting for Anything

Beomyoung Kim, Chanyong Shin, Joonhyun Jeong, Hyungsik Jung, Se-Yun Lee, Sewhan Chun, Dong-Hyun Hwang, Joonsang Yu

机构 * NAVER Cloud, ImageVision（NAVER云，ImageVision）

Comments ICCV 2025 (Highlight)

2408.04631 2025-08-29 cs.CV cs.AI

Puppet-Master: Scaling Interactive Video Generation as a Motion Prior for Part-Level Dynamics

Ruining Li, Chuanxia Zheng, Christian Rupprecht, Andrea Vedaldi

机构 * Visual Geometry Group, University of Oxford（牛津大学视觉几何组）

Comments Accepted at ICCV 2025. Project page: https://vgg-puppetmaster.github.io/